Computer Science Student · Full-Stack Developer · AI Researcher — Building reliable, well-architected software and researching applied AI — from productivity platforms to graph databases to deep learning for mental health.
I'm a Computer Science student at UET Lahore, graduating in 2027, working across the full stack, mobile, and applied AI. My projects span production-style platforms — automating document workflows and AI insights in WorkPulse — to deeper systems work like building a graph database engine from scratch, and research into deep learning for mental health detection. I care as much about how a system is architected — its data flow, its failure modes, its consistency guarantees — as I do about how it looks and feels to use. That instinct comes from moving between disciplines: shipping full-stack products, building embedded IoT systems, and writing research papers, all in the same academic year.
Grouped by where they sit in a system — from interface to infrastructure.
From production-style automation to research prototypes and embedded hardware — each one taught me something the others didn't.
AI-powered productivity monitoring platform
A full-stack productivity monitoring platform that automates document intake, tracks work activity, and delivers AI-driven insights. Built with a Python/Flask backend, Supabase for data and auth, n8n for workflow automation, and EasyOCR for document parsing.
Android multi-game platform, 7 games in one app
A feature-rich Android multi-game platform bringing together seven casual games — Memory Match, Neon Dash, Piano Tiles, Photo Puzzle, Quiz Challenge, Rapid Tap, and Word Quest — in a single app, with authentication, cloud-synced profiles, and global leaderboards.
Real-time object detection and tracking desktop app
An AI-powered desktop application for real-time object detection and tracking using the YOLOv8 deep learning model, with an intuitive GUI for analyzing images, videos, and live webcam feeds with high accuracy and speed.
Graph-native data engine with visual querying
A from-scratch graph database engine with a Streamlit front end and FastAPI backend, supporting ACID-compliant transactions, Cypher-like query syntax, and interactive graph visualization.
LSTM-based deep learning research
A research project applying LSTM-based deep learning to detect indicators of depression from social media text, combining NLP preprocessing with sequence modeling to classify linguistic and behavioral patterns.
Spreadsheet engine built on core DSA
A spreadsheet application built entirely on fundamental data structures — graphs for dependency tracking, trees for parsing, and stacks/queues/linked lists for evaluation — supporting live formula computation.
This research investigates the use of Long Short-Term Memory (LSTM) networks to detect linguistic and behavioral indicators of depression in social media text. By combining NLP preprocessing techniques with sequence-based deep learning, the study aims to classify posts as depression-indicative or non-indicative, contributing to early-detection tooling for mental health awareness.
B.Sc. Computer Science · Lahore, Pakistan
Currently in progress · CGPA 3.4/4.0
FSc Pre-Engineering · Lahore, Pakistan
Completed
Zenodo (DOI: 10.5281/zenodo.21265208) · 2026
University of Engineering and Technology (UET), Lahore · 2023 — Present
Live contribution data can be wired up via the GitHub GraphQL API — this is a placeholder graph until that's connected.
Wire this up to the GitHub GraphQL API (contributionsCollection) for live data, or embed a service like github-readme-stats.
Whether you're hiring, supervising research, or just want to compare notes on graph databases — I'd like to hear from you.